The Impacts of Economic Policy Uncertainty on Trade Flow

ABSTRACT

This paper examines the impacts of economic policy uncertainty (EPU) on the trade flow of 143 countries, considering the EPU of exporter and importer countries and the income heterogeneity of the sample countries for over 2000–2019. Besides, it considered the mediating role of regional trade agreements (RTAs) in the EPU-trade nexus. Our results of the two-step Heckman sample selection model imply that the intensive and extensive margins of trade are adversely affected by the exporter and importer countries’ EPU. However, the negative effects of EPU of exporter countries on trade performance turn to positive when the mediating role of the RTAs is controlled. Also, the counterfactual simulation analysis is conducted to compute illustrative distance equivalents of improving the EPU to the average performing country in the sample. Finally, the marginal effect of EPU with respect to GDP per capita results indicates that the higher levels of per capita income seem to decrease the negative effect of EPU on trade. Our findings are robust after we run sensitivity analysis using the Poisson pseudo-maximum likelihood (PPML) to control heterogeneity and zero-valued observations issues. Finally, based on the findings, policy implications are forwarded.

KEYWORDS:

• Economic policy uncertainty (EPU)
• trade flow
• structural gravity model

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Notes

1. Using MRT captures the role of country size because trade barriers have a large impact on small countries, which typically trade a large proportion of their output internationally (Portugal-Perez and Wilson, 2012). So, we replace bilateral variables that account for bilateral trade costs by MRT in the model using the following Baier and Bergstrand (2009) methods BY indexing (i,b,c) for the reporter and (j,e,f) for partner countries. $\mathrm{M}\mathrm{R}\mathrm{T}\mathrm{l}\mathrm{n}{\mathrm{X}}_{\mathrm{i}\mathrm{j}\mathrm{t}}=\mathrm{l}\mathrm{n}{\mathrm{X}}_{\mathrm{i}\mathrm{j}\mathrm{t}}-\left\{\sum _{e\notin i}^{\hspace{0.17em}}{\theta }_{i}ln{X}_{iet}+\sum _{b\notin j}^{\hspace{0.17em}}{\theta }_{j}ln{X}_{bjt}-{\sum }_c^{\hspace{0.17em}}{\sum }_f^{\hspace{0.17em}}{\theta }_j{\theta }_{i}ln{X}_{cft}\right\}$. Where X is bilateral variables accounting for bilateral trade costs, θ_i = Y_i/Y_T and θ_j = Y_j/Y_T, Y_i = GDP_it, Y_T = GDP_world.

Additional information

Funding

This work was supported by the General Project of the Social Science Fund [20BSH103]; National Natural Science Foundation [72172129].